Young people's preferences for family planning service providers in rural Malawi: a stated preference data set - codebook

Permanent Identifier:

https://doi.org/10.17037/DATA.5

Data Description

Data collected as part of a discrete choice experiment (DCE) conducted among young people aged 15-24 in seven villages in rural Malawi. The aim of the DCE was to explore stated preferences for family planning service providers. Data collection took place in 2012. The overall DCE included 20 choice tasks which were blocked into 4 groups of 5 using an efficient design algorithm in Ngene. Selected sociodemographic characteristics are included.

Discrete Choice Experiment Dataset codebook

Variable Name STATA Label Creator-provided description Variable Type
P_ID Participant ID number Unique Participant ID number long
D_Ques Choice task number from efficient design Choice task number from unblocked efficient design. Numeric
Q_Ver Questionnaire version A or B Two versions of the questionnaire were used. In Version A the choice of family planning service provider DCE followed a DCE related to the design of outreach services.  In Version B the the choice of family planning service provider DCE wa spresented first, followed by the DCE related to the design out outreach services. This variable indicates version A or B. String
Q_VerA Version A  (1=A 0=B) Dummy variable indicating whether  questionnaire version A was used. 1=A. 0=B. long
D_Blk Block number from efficient design The experimental design included 20 choice tasks which were divided into 4 blocks of 5 choice tasks each. This variable indicates the block number from 1-4. Numeric
D_Alt Choice alternative (1=gov  2=private  3=outreach 4=CBDA 5=none) The DCE included 4 labelled alternatives and an opt-out. This variable indicates the alternative associated with the attributes describing a particular choice task. 1=Government Service.  2=Private Clinic. 3=Outreach Service. 4=Community based distributor (CBDA). 5=None/Neither Numeric
C_Resp Alternative selected by repondent Indicates the alternative that a respondent choice for a given choice task Numeric
C_Choice Coded resp variable 1=alternative selected 0=alternative not selected Dummy coded response variable indicating the alternative that a respondent. For each choice task there will be one chosen alternative and 4 unchosen alternatives. Chosen alternative=1. Unchosen alternatives=0. Numeric
D_Cset Number of alternatives within each choice set Each respondent was presented with 5 choice tasks. This is referred to as the choice set and takes a fixed value of 5 for all observations. This variable is required for analysis to indicate the chosen and non-chosen alternatives. Numeric
D_NChoices Number of choice tasks presented to each respondent This variable indicates the number of choice tasks that each respondent was presented with. Each respondent was presented with 5 choice tasks; this variable therefore takes on a fixed value of 5. This is required for analysis to indicate that there are multiple observations from each respondent. Numeric
D_BlkQs Choice task number within block Indicates the choice task number within each block of 5 choice tasks. Numeric
A_Dist Attribute - distance from home to facility in kilometers Describes alternative family planning service providers in terms of distance in kilometers from repondent's home. Levels are alternative specific for government and private services and fixed for outreach and CBDA services. Numeric
A_Freq Attribute - frequency that services are available Describes alternative family planning service providers in terms of the frequency that services are available.  Fixed values for government, private and CBDA services (8am-5pm Monday-Friday). Outreach service varies from once per month to once every two months. Numeric
A_Drug Attribute - availability of FP commodities Describes alternative family plannig service providers in terms of availability of family planning commodities. Effects coded. -1 indicates that commodities are sometimes not available or not all methods are available at the facility. 1 indiccates that all methods are available. Value fixed to 1 for outreach service. Numeric
A_Svcp Attribute - service provider attitude Indicates the attitude of the family planning service provider for a given alternative. Effects coded. -1 indicates an unfriendly provider who may be rude or scold clients. 1 indicates a friendly and open provider. Value fixed at 1 for outreach service. Numeric
A_Wait Attribute - waiting time to see a service provider Indicates the waiting time in minutes to see a family planning service provider for a given alternative.  Fixed to 0 for CBDA alternative since this service is provided in the individual's home and there is no need to wait once the provider arrives. Numeric
A_Price Attribute - price Indicates the price of family planning services for a given alternative. Fixed at 0 for government service alternative as government services were always provided free of charge in this setting. Numeric
P_HHID Participant household ID Indicates the household ID for the participant. Used for tracing participants in the field. Randomisation at individual level means that there may be more than one respondent per household. Integer
P_VID Participant village ID Indicates the participants' village of residence.  Seven villages were included in the survey. String
P_Age Participant Age Indicates respondents' age as a continuous variable in years. double
P_Old Participant age category (15-19=0 20-24=1) Dummy variable indicating respondent's age category. 15-19=0. 1=20-24. Numeric
P_Gender Participant gender (1=Female 2=Male) Indicates respondent gender. Female=1. Male=2. Numeric
P_Fem Participant is Female (Male=0 Female=1) Dummy variable indicating whether respondent is female.  Female=1. Male=0. Numeric
P_Edu Currently attending school (1=Yes 0=No ) Dummy variable indicating whether the respondent is currently attending school. Yes=1. No=0. Numeric
P_EduLev Participant highest level of educational attainment Indicates repondents' highest level of schooling completed. Lower primary (standard 1-4)=1. Upper primary (standard 5-8)=2. Lower secondary (form 1-2)=3. Upper secondary (form 3-4)=4. University/College=5. Technical training=6. Other=96. Numeric
P_Empl Paid work in the last 12 months (0=No 1=Yes) Dummy variable indicating whether respondent engaged in paid employment in the 12 months preceding the survey. Yes=1. No=0. Numeric
P_Rltn Relationship in the last 12 months (0=No 1=Yes) Dummy variable indicating whether respondent was in a relationship in the 12 months preceding the survey. This includes marriage and boyfriend/girlfriend relationships. No=0. Yes=1. Numeric
P_EvrSx Has ever had sex (0=No 1=Yes) Dummy variable indicating whether respondent has ever had sexual intercourse. No=0. Yes=1. Numeric
P_Sex12 Had sex in the last 12 months (0=No 1=Yes) Dummy variable indicating whether respondent has had sex in the 12 months preceding the survey. This is an indicator of possible need for fmaily planning. No=0. Yes=1. Numeric
P_FPuse FP use in the last 12 months (0=No 1=Yes) Dummy variable indicating whether respondent used any modern method of family planning (contraception) in the 12 months preceding the survey. Includes condoms, oral contraceptives, injectables, IUD and emrgency contraception. No=0. Yes=1. Numeric
P_FPprice Amount participant paid for FP at the last place attended Indicates the amount that respondents who had used a modern method of family planning in the 12 months preceding the survey paid for the last (most recent) family planning method they received. Amounts in Malawi Kwacha. At the time of the survey in May-June 2012, 50 Malawi Kwacha was equal to approximately USD$0.20. double
P_DKprice Participant doesnt know how much paid for last FP (0=Knows 1=Doesn't know) Dummy variable indicating whether respondents who had obtained a modern method of family planning in the 12 months preceding the survey were able to recall how much they paid for the last (most recent) family planning method they received. Participant knows the price paid=1. Participant doesn't know=0. Numeric
R_TooManyQ Review - There were too many choice tasks (0=No 1=Yes) Dummy variable indicating whehter the respondent thought that there were too many choice tasks for the choice of family planning service provider DCE. Yes=1. No=0. Numeric
R_Boring Review - The choice tasks were boring (0=No 1=Yes) Dummy variable indicating whehter the respondent thought that the choice tasks for the choice of family planning service provider DCE were boring. Yes=1. No=0. Numeric

Malawi DCE Experimental Design Dataset codebook

Variable Description
Choice situation Choice task number from efficient design
gov.g_dist Distance value for government service alternative
gov.g_drug Availability of family planning commodities values for government service alternative
gov.g_svpv Service provider attitude for government service alternative
gov.g_wait Waiting time for government service alternative
pvt.p_dist Distance value for private clinic alternative
pvt.p_drug Availability of family planning commodities values for private clinic alternative
pvt.p_svpv Service provider attitude for private clinic alternative
pvt.p_wait Waiting time for private clinic alternative
pvt.p_price Price for private clinic alternative
out.o_wait Waiting time for outreach service alternative
out.o_freq Frequency with which outreach sevrices are offered in respondents' community
out.o_price Price for outreach service alternative
cbda.c_drug Availability of family planning commodities values for community based distribution alternative
cbda.c_svpv Service provider attitude for community based distribution alternative
cbda.c_price Price for community based distribution alternative
Block Block number. The experimental design included 20 choice tasks which were divided into 4 blocks of 5 choice tasks each. This variable indicates the block number from 1-4.